According to a prevalent memory model, episodic memory traces are stored in the same sensory-specific corti- cal regions that processed the event during encoding, such as the ventral occipito-temporal cortex (VOTC) for visual information. During memory retrieval frontoparietal regions interact with medial temporal lobe (MTL) regions, giving rise to the reactivation of distributed memory traces and the experience of remembering. Abundant data supports reactivation theory, including fMRI evidence that some of the same brain regions activated during encoding are reactivated during retrieval. According to the differentiation hypothesis, the amount of mnemonic details recovered during reactivation increases over time; memories start vague but can grow to be clearer (more differentiated). During recognition tests, differentiation for some memories reaches the familiarity level, in which one knows the event occurred in the past but cannot retrieve its contextual de- tails (e.g., when, where, etc.). For other memories, differentiation continues further to the recollection level, wherein contextual details can be identified. This project investigates two fundamental aspects of reactivation theory and the differentiation hypothesis. First, modality-specific reactivation has been detected as early as 100ms after the presentation of a retrieval cue, with the latency of this early lateralized reactivation effect (eLRE) precluding a role for the hippocampus (>200ms), an MTL structure critical for the conscious episodic retrieval of recently learned information. This suggests the eLRE reflects low-level sensory memory traces that are formed implicitly during encoding and do not require hippocampal processing to occur at retrieval. Study 1 uses high temporal resolution EEG to investigate how the eLRE contributes to later familiarity (Frontal-Famili- arity Effect, ~400 ms) and recollection (Parietal-Recollection Effect, ~600ms). Second, the temporal and network dynamics of longer-latency modality-specific reactivation in the VOTC are not well understood. Study 2 simultaneously records fMRI and EEG to test the hypotheses that more differentiated memories arise at later latencies and are more cortically distributed than less differentiated memories. Additionally, within MTL, famili- arity has been linked to the rhinal cortex (RC), and recollection to the hippocampus (HC). In frontoparietal cortex, familiarity has been associated with the dorsal frontoparietal control network (FPN), and recollection with ventral-parietal cortex (VPC). Using data from Study 2, I will test the prediction that familiarity-level VOTC reactivation will covary with RC and FPN activity and the Frontal-Familiarity effect, whereas recollection- level VOTC reactivation will covary with HC and VPC activity and the later Parietal-Recollection effect. In sum, using a sophisticated combination of methods, the proposed studies will reveal the spatiotemporal dynamics of reactivation and its relationship to the mechanisms and theoretical underpinnings of memory. The findings will also have important clinical implications for understanding memory disorders in psychiatric patients, while also helping the applicant acquire critical knowledge and skills for a career in cognitive neuroscience.